Storage system manufacturers have developed mechanisms to make their systems “highly available” (HA) in the event that any one or more hardware components within a given storage system fails. For example, when a hardware component fails, another component “steps in” to takes its place, and a server or consumer of the data is able to continuously access the data with little or no interruption to data access.
One conventional HA mechanism includes connecting at least two independent controllers to a Serial Attached Small Computer System Interface (SAS) or Fibre Channel (FC) disk drive at the same time. Such independent controllers are active or passive, in which only one controller has access to the SAS or FC disk drive at any given time. Another conventional HA mechanism includes mirroring data from a first storage system to a second storage system via a network link. In this scenario, each of the storage systems maintains a copy of the data. Such conventional failover technologies are inefficient and/or require specialized disk drives.
The above-described deficiencies of today's failover environments and related technologies are merely intended to provide an overview of some of the problems of conventional technology, and are not intended to be exhaustive, representative, or always applicable. Other problems with the state of the art, and corresponding benefits of some of the various non-limiting embodiments described herein, may become further apparent upon review of the following detailed description.